Real time measures of prestin charge and fluorescence during plasma membrane trafficking reveal sub-tetrameric activity

Abstract

Prestin (SLC26a5) is the outer hair cell integral membrane motor protein that drives cochlear amplification, and has been described as an obligate tetramer. We studied in real time the delivery of YFP-prestin to the plasma membrane of cells from a tetracycline-inducible cell line. Following the release of temperature block to reinstate trans Golgi network delivery of the integral membrane protein, we measured nonlinear capacitance (NLC) and membrane fluorescence during voltage clamp. Prestin was delivered exponentially to the plasma membrane with a time constant of less than 10 minutes, with both electrical and fluorescence methods showing high temporal correlation. However, based on disparity between estimates of prestin density derived from either fluorescence or NLC, we conclude that sub-tetrameric forms of prestin contribute to our electrical and fluorescence measures. Thus, in agreement with previous observations we find that functional prestin is not an obligate tetramer.

Figure 1. Prestin delivery to the cell surface can be monitored in real time by NLC measurements.

A) For initial experiments, tet-inducible HEK293 cells were incubated with 1.0 µg/ml tetracycline for 30 min at 37°C prior to whole cell recording in normal growth media plus tetracycline at room temperature (24°C). In this case, intracellular (pipette) solution was (in mM) 128 KCl, 5 MgCl₂, 0.5 CaCl₂, 5 EGTA, pH7.28. Cell membrane capacitance traces were recorded every 5 minutes, and the time stamp on traces indicates the time following tetracycline. Prestin NLC clearly develops over time. All traces are subtracted with the 43 minute trace. B) Synchronized transient delivery of newly synthesized prestin to the plasma membrane from TGN occurs after release of T-block. Freshly plated tet-inducible HEK293 cells on cover slips were allowed to settle for 3 hours at 37°C before incubation with 1.0 µg/ml tetracycline for 60 min at 21°C for protein synthesis. Cells were then transferred to a bath solution composed of normal growth media plus tetracycline supplemented with 4 mM 4-AP and 5 mM TEA for whole-cell patch clamp. Intracellular (pipette) solution is given in Methods. Cell membrane capacitance traces were taken every 2 minutes, and the release of temperature block was initiated by increasing the temperature of the bath solution after whole-cell configuration was established and the cell was detached using the pipette. In order to emphasize the increase over time, NLC traces are subtracted with 0 time trace. Prestin trafficking from trans-Golgi network to the plasma membrane occurs quickly. C) Normalized NLC change during transient delivery of prestin to the plasma membrane following release of T-block. NLC changes either in Peak NLC or in fitted Q_max were averaged (+/− se, n = 3–8 cells) for each time point as measured in B. Normalized NLC data were fit with a single exponential, τ being 6.4 minutes. The plateau likely reflects the depletion of membrane-bound prestin molecules from the trans-Golgi complex.

Figure 2. Prestin delivery to the cell surface can be monitored in real time by fluorescence measurements.

A) A phase-contrast image and consecutive YFP fluorescence photos taken every two minutes, simultaneous with NLC measurements as in Figure 1 . Exposure conditions: SPOT CCD camera, 8 second exposure, 16× NDF, 40× lens. While there is a clear photo-bleaching effect intracellularly, the fluorescence intensity on the cell membrane is kept near constant, indicating an added component from trafficking. B) Measurement of intracellular integrated fluorescence density (open diamonds) and that of the membrane (open circles) as indicated by the ROI in the photo insert. Following bleach correction according to Equations (2–5) membrane fluorescence shows an increase over time (filled circles). C) Membrane fluorescence changes for each time point as measured above were averaged (+/− se, n = 14 cells) and normalized. Normalized NLC data were fit with a single exponential, τ being 7.9 minutes. As with NLC, the plateau likely reflects the depletion of membrane-bound YFP-prestin molecules from the trans-Golgi complex.

Figure 3. Correlation of membrane surface fluorescence and NLC.

Data from Fig. 1C and Fig. 2C are plotted against each other. The high correlation (r² = 0.983) between the two confirms that either can be used to monitor kinetics of prestin delivery to the plasma membrane following release from T-block.

Figure 4. Estimates of prestin monomers deleiverd to the membrane following release from T-block.

A) Prestin molecules transported to cell membrane visualized as diffraction-limited fluorescence spots consistent in size with single molecules (Hallworth and Nichols, J Neurophysiology, 2011). Membrane pieces observed with fluorescent microscope under the same condition as described in Fig. 2 for monitoring trafficking. Scale bar 10 µm. B) Enlarged view for particle picking for fluorescence quantification. Uniform box size 4×4 square pixels corresponding to 760×760 nm² area, very close to the box size used by Hallworth et al. (750×750 nm²). Bright, larger spots likely representing multiple particles were excluded. Integrated fluorescence of each spot was quantified after global background correction, the same way as done in Fig. 2 . C) Particle intensity distribution. Data were automatically binned and fitted in Origin. The fact that the regression line does not pass through the origin may indicate that non-functional fluorescent monomers contributed to membrane fluorescence. D) Estimates of monomer deposition into the membrane obtained using NLC or fluorescence measures (see Methods). Grey dotted lines depict +/− standard deviation. Note discrepancy, with NLC estimates giving greater than an order of magnitude increase in numbers of monomers.